# KellerJordan/Muon

Muon is an optimizer for hidden layers in neural networks

Repository: https://github.com/KellerJordan/Muon
Canonical: https://ross.abutalabs.com/products/kellerjordan-muon
Language: Python
License: MIT
License Family: permissive
Last push: 2026-05-24T18:39:02+00:00

## Health v2 (maintenance only)
Score: 59/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 84, release rhythm 35, longevity 47
- inputs: {"age_days": 662, "days_push": 101, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 2801, forks 129 (observed 2026-08-28T04:07:22.558958+00:00)

## What it is
Muon is a PyTorch optimizer for the hidden layers of neural networks, based on orthogonalized momentum updates via Newton-Schulz iteration. It is designed to be combined with AdamW for embeddings, heads, and biases, and has set training speed records on GPT-2 and CIFAR-10 benchmarks.

## Use cases
- train transformers faster than with AdamW
- optimize hidden layer weights of a neural network
- speed up GPT-2 style LLM pretraining on a budget
- set CIFAR-10 training speed records
- replace AdamW for hidden weights in a PyTorch model
- scale learning rate across model sizes with muP-style scaling

## When to choose
- you are training transformers or convnets in PyTorch and want faster convergence per FLOP
- you want a drop-in optimizer for hidden weights alongside AdamW
- you are running training speedrun experiments like modded-nanogpt

## When to avoid
- you need a battle-tested general-purpose optimizer for all parameter types
- you are not using PyTorch
- you need embeddings, heads, and biases optimized by the same optimizer

## Facets
- artifact type: library
- maturity: active
- function: machine-learning, llm-training, deep-learning
- domain: machine-learning, deep-learning, large-language-models
- platform: python
- tags: optimizer, pytorch, newton-schulz, training-speedrun, muon

## Member repositories
- KellerJordan/Muon (main) score 59

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:07:22.558958+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T08:14:33.626062+00:00, confidence not recorded.
  - readme: https://github.com/KellerJordan/Muon (fetched 2026-08-28T04:07:22.558958+00:00, sha bb973148d70b)
- Data as of 2026-08-30T08:39:29.467469+00:00.